Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

eff.org

Post reply on HN
Print view View original

111–120 of 170 posts

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#111

Post by rayiner »

DannyBee wrote: post #105
rayiner wrote: post #102

Earlier quoted context omitted.

> You know it's not the only question, and not even the most relevant one (which is whether the structure, sequence, and organization part of an API is the copyrightable part, or whether the literal implementation only is protected). To me, an "API" is just the "structure, sequence, and organization part," distinct from the implementation. I.e. you can't take a C header file and delete the parameter names, and replac…

" I.e. you can't take a C header file and delete the parameter names, and replace all the types with typedef'ed names and get a different API" Of course, if you did so, the fact that you ended up with the same API means you have removed any expressive portions, and are left with purely functional constructs and short names. ". And under 17 USC 102, how you answer that question is to look at whether that is original a…

> Of course, if you did so, the fact that you ended up with the same API means you have removed any expressive portions, and are left with purely functional constructs and short names.

That presupposes that there is nothing expressive in the structure of the API itself, beyond the names. But the combination and structure of calls in an API is an expression in and of itself, as evidenced by the fact that the same functionality (e.g. file access) can be expressed in many different ways.

> Sorry, but no. The fact that something is original and creative does not, by itself, make something copyrightable in any way, shape, or form.

I didn't say it was the only requirement for copyrightability. There's obviously the exclusions in 17 USC 102(b). But in addition to the "work of authorship" and "fixation" requirements in 17 USC 102(a), which I assume aren't in dispute, originality is what matters.

> It 100% is not indicative of this. It is hard work to compile address lists and phone numbers into the white pages.

Ditch-digging requires a lot of work too, and I'm not asserting that is creative, but I do think API design is a creative process that requires a lot of careful thought.

> However, as a concept, it is one completely and totally ineligible for copyright protection.

Sure, the concept of passing around file descriptors as small integers is not protectable in the abstract. But it's an expressive choice that, when weaved into an API, takes it beyond purely functional into a creative expression.

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#113

Post by DannyBee »

rayiner wrote: post #111
DannyBee wrote: post #105

Earlier quoted context omitted.

" I.e. you can't take a C header file and delete the parameter names, and replace all the types with typedef'ed names and get a different API" Of course, if you did so, the fact that you ended up with the same API means you have removed any expressive portions, and are left with purely functional constructs and short names. ". And under 17 USC 102, how you answer that question is to look at whether that is original a…

> Of course, if you did so, the fact that you ended up with the same API means you have removed any expressive portions, and are left with purely functional constructs and short names. That presupposes that there is nothing expressive in the structure of the API itself, beyond the names. But the combination and structure of calls in an API is an expression in and of itself, as evidenced by the fact that the same func…

"That presupposes that there is nothing expressive in the structure of the API itself, beyond the names. But the combination and structure of calls in an API is an expression in and of itself, as evidenced by the fact that the same functionality (e.g. file access) can be expressed in many different ways."

By this argument, everything ever, including my shopping lists are creative expression and copyright eligible.

Copyright protection in software has never extended this far, and your argument is not particularly convincing (at least to me :P)

"Sure, the concept of passing around file descriptors as small integers is not protectable in the abstract. But it's an expressive choice that, when weaved into an API, takes it beyond purely functional into a creative expression."

We're going to have to agree to strongly disagree.

By your abstract arguing, almost everything can be found to be expressive as long as it's "an expressive choice". This would extend copyright far into the realm of methods and processes.

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#114

Post by whafro »

talideon wrote: post #60
jobu wrote: post #48

Earlier quoted context omitted.

Book Titles are not copyrightable, even though they are arguably the most important part of a creative work. While APIs are not a clear analog to titles, they are a huge exception to the copyright law, and it seems like APIs need this exception as well.

They can, however, be trademarked.

Sure, but only once the title has attained secondary meaning over and above the words in the title. Basically, it has to be beyond the level of a bestseller. Most book titles don't come anywhere close.

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#115

Post by hyperpape »

AnthonyMouse wrote: post #97
steveklabnik wrote: post #87

Earlier quoted context omitted.

It also doesn't mean that they're not still Google employees.

What are you proposing then? A conflict of interest requirement for amici? Nobody motivated to write one would ever be allowed to.

Like the earlier post said, amicus curiae briefs are filed on behalf of people who are not parties to a case. They may have an interest, but it can't be "we're suing/being sued in this particular case." Maybe it's different for employees of a firm, but Google is certainly a party to the case.

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#116

Post by orangecat »

kybernetyk wrote: post #13
DevX101 wrote: post #3

Although these issues are very important, something feels wrong about having the Supreme Court justices (avg. age ~70) rule on tech matters like this. We need a new legal system for arbitrating disputes in technical disciplines overseen by a panel of technical experts. Or perhaps a trial by jury where the selection criteria is employment in the relevant discipline for at least 10 years (academia or corporate).

> avg. age ~70 Why is that a problem? It's certainly possible that the judge was an avid home computer enthusiast in the 1980s and knows about tech. Old age doesn't automatically mean technological incompetence anymore.

Old age doesn't automatically mean technological incompetence anymore.

For example, Judge Alsup was born in 1945 and demonstrated a perfect understanding of the issues in his ruling. As I recall he even taught himself Java programming during the trial.

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#117

Post by toyg »

wmf wrote: post #49
musesum wrote: post #42

Earlier quoted context omitted.

I find "30. John Gage" interesting in that he was Chief Scientist at Sun. I always thought that Google's Dalvik was a hack around Sun's business model for Java, in bypassing the need to license JVM. The fact that John Gage signed the brief seems to imply that he considers Google's hack to be less ominous than Oracle's attempt to change the game.

The Java class/jar file formats are inherently slow and Dalvik's dex code fixes that problem. I don't think the licensing hack would have worked since you can't escape patents by making syntax changes.

Well, that is what the courts are discussing at the moment. It may or may not work, but the origins of Dalvik as a "licensing hack" are well-known.

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#118

Post by Fargren »

jobu wrote: post #48
spullara wrote: post #43

In a world where people copyright the laws you are supposed to follow I can't imagine they won't let people copyright an API that absolutely is a creative work. Instead we should accept that they can be under copyright but make implementing and using them explicitly fair use.

Book Titles are not copyrightable, even though they are arguably the most important part of a creative work. While APIs are not a clear analog to titles, they are a huge exception to the copyright law, and it seems like APIs need this exception as well.

So if I were to write a book named A Game of Thrones, under the pen name George R. R. Martin, would it be permissible by law to publish it?

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#119

Post by shmerl »

sillysaurus3 wrote: post #69

Quick question: Why can't an API be copyrighted? My understanding is that the purpose of a copyright is to protect a specific arrangement of ideas. A piece of music, a book, a painting. All of those things are an arrangement, and the arrangement itself is what's protected by copyright, not the physical object that was created. So, given that an API is an arrangement of ideas, and that copyright apparently protects ar…

Simple answer - it's too generic of a level to allow copyrighting it. It ends up in vendor lock-in and anticompetitive behavior. Same reason abstract ideas aren't allowed to be copyrighted - it gives too much monopoly to those who would claim them.

Re: Computer Scientists Ask Supreme Court to Rule APIs Can’t Be Copyrighted

#120

Post by throwawaykf05 »

jrochkind1 wrote: post #88

Many in this thread seem to be wondering about the legal argument itself -- that API's being copyrightable is inconsistent with existing pretty well-established law. Which is, yes, a different thing than arguing about whether it's "a good idea" for society or innovation or whatever. The EFF summary focuses on the "good idea" thing, but the actual amicus brief addresses both -- although I bet you can find even more of…

I don't think the cited case law for exemptions for interoperability in interfaces applies to this case. For instance, the Sega and Sony cases apply for, as far as I can tell, computer systems interoperating with each other. Lotus applies to humans operating software applications running on computer systems. APIs on the other hand, are mainly for humans to communicate with each other when building systems that interoperate. As such, APIs are primarily a means of expression between humans - from those who provide the interface, to those who use it. So from a copyright perspective, I don't see how binary interoperability between systems or the GUI layouts can compare to textual expression between humans.

Also, as others have mentioned, the whole "method of operation" argument is also a dangerous slope to be slipping down. What is all code anything other than a "method of operation" of the hardware it runs on?

I've said it before, but copyright is not really a good mechanism for IP protection of code. Software was shoehorned under copyright because 1) other mechanisms were not available and 2) there is sometimes creative expression involved in writing code. It's only the potential of creative expression what grants copyright protection to any written code. By that standard, APIs are no different from any other code.

Post reply on HN